(OBIFS) isotropic image analysis for improving a predicting agent based systems

• A hybrid forecasting system for predicting the future situation of the forest fires.
• The system presents a distributed multi-agent architecture.
• We use a buffer operator for case-base creation.
• The use of the buffer operator improves the quality of the date used by the system.
• The model outperforms the results obtained by others previous tested methods.

In this interdisciplinary study a novel hybrid forecasting system is presented, in which an isotropic buffer operator is applied for case-based creation within the structure of the organization-based multi-agent system. Commonly used as an image analysis technique by commercial Geographic Information Systems (GIS), the buffer operator in this particular system calculates the area of a forest fire for prediction and visualization tasks. The use of the buffer operator improves the quality of the data used by the system and in consequence the quality of the results obtained. The system has been successfully tested using real historical data on forest fires evolution, by generating accurate predictions.